deepseek-ai/DeepSeek-OCR
Image-Text-to-Text
•
3B
•
Updated
•
1.34M
•
2.26k
Vola Delta
hf-delta
AI & ML interests
LLM
Recent Activity
liked
a model
11 days ago
deepseek-ai/DeepSeek-OCR
liked
a model
24 days ago
nineninesix/kani-tts-370m
liked
a model
25 days ago
deepseek-ai/DeepSeek-V3.2-Exp
Organizations
None yet
liked
a
model
11 days ago
liked
a
model
24 days ago
liked
2
models
25 days ago
upvoted
a
collection
about 1 month ago
liked
2
models
3 months ago
liked
a
model
5 months ago
reacted to
codelion's
post with 🔥
5 months ago
Post
2364
Introducing AutoThink: Adaptive reasoning for LLMs that improves performance by 43% on reasoning benchmarks!
Instead of using fixed thinking budgets, AutoThink:
- Classifies query complexity (HIGH/LOW) using adaptive classification
- Dynamically allocates thinking tokens based on complexity
- Uses steering vectors derived from Pivotal Token Search to guide reasoning patterns
Results on DeepSeek-R1-Distill-Qwen-1.5B:
- GPQA-Diamond: 31.06% vs 21.72% baseline (+9.34 points)
- MMLU-Pro: 26.38% vs 25.58% baseline (+0.8 points)
- Uses fewer tokens than baseline approaches
Works with any local reasoning model - DeepSeek, Qwen, Llama, custom models. The technique combines our research on Pivotal Token Search (PTS) implementation and adaptive classification frameworks.
Paper: AutoThink: efficient inference for reasoning LLMs
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5253327
Code and examples:
https://github.com/codelion/optillm/tree/main/optillm/autothink
PTS implementation and technical details:
https://github.com/codelion/pts
https://huggingface.co/blog/codelion/pts
Adaptive classifier framework:
https://github.com/codelion/adaptive-classifier
Would love to hear your thoughts on adaptive resource allocation for LLM reasoning! Have you experimented with similar approaches?
Instead of using fixed thinking budgets, AutoThink:
- Classifies query complexity (HIGH/LOW) using adaptive classification
- Dynamically allocates thinking tokens based on complexity
- Uses steering vectors derived from Pivotal Token Search to guide reasoning patterns
Results on DeepSeek-R1-Distill-Qwen-1.5B:
- GPQA-Diamond: 31.06% vs 21.72% baseline (+9.34 points)
- MMLU-Pro: 26.38% vs 25.58% baseline (+0.8 points)
- Uses fewer tokens than baseline approaches
Works with any local reasoning model - DeepSeek, Qwen, Llama, custom models. The technique combines our research on Pivotal Token Search (PTS) implementation and adaptive classification frameworks.
Paper: AutoThink: efficient inference for reasoning LLMs
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5253327
Code and examples:
https://github.com/codelion/optillm/tree/main/optillm/autothink
PTS implementation and technical details:
https://github.com/codelion/pts
https://huggingface.co/blog/codelion/pts
Adaptive classifier framework:
https://github.com/codelion/adaptive-classifier
Would love to hear your thoughts on adaptive resource allocation for LLM reasoning! Have you experimented with similar approaches?
- 5 replies
liked
a
model
7 months ago
liked
4
models
8 months ago
liked
a
model
10 months ago
upvoted
a
collection
10 months ago
upvoted
a
collection
11 months ago
liked
2
models
11 months ago
liked
a
model
12 months ago
